app.py

import streamlit as st
from streamlit_chat import message
from langchain_openai import ChatOpenAI
from langchain.chains import ConversationChain
from langchain.chains.conversation.memory import (ConversationBufferMemory, 
                                                  ConversationSummaryMemory, 
                                                  ConversationBufferWindowMemory
               
                                                  )

if 'conversation' not in st.session_state:
    st.session_state['conversation'] =None
if 'messages' not in st.session_state:
    st.session_state['messages'] =[]
if 'API_Key' not in st.session_state:
    st.session_state['API_Key'] =''

# Setting page title and header
st.set_page_config(page_title="ChatMate: Your Professional AI Conversation Partner Solution", page_icon=":robot_face:")
st.markdown("<h1 style='text-align: center; color: navy;'>ChatMate</h1>", unsafe_allow_html=True)
st.markdown("<h4 style='text-align: center;'>A cutting-edge language model</h4>", unsafe_allow_html=True)
st.markdown("<p style='text-align: right'>By <a href='https://entzyeung.github.io/portfolio/index.html'>Lorentz Yeung</a></p>", unsafe_allow_html=True)

st.markdown("<p style='text-align: left;'>I am capable of recalling previous parts of our conversation, such as remembering your name if you share it with me.</p>", unsafe_allow_html=True)
st.session_state['API_Key']= st.text_input("First, to get it work, put your OpenAI API Key here please, the system will enter for you automatically.",type="password")
st.markdown("<p style='text-align: left;'>Then Tell me how I can help:</p>", unsafe_allow_html=True)



# API Keys
# st.sidebar.text_input() will automatically update st.session_state['API_Key'] with the input value whenever the user types into the field. 
st.sidebar.title("Introduction")
st.sidebar.markdown("""
ChatMate is an advanced conversational AI interface, expertly crafted to demonstrate the fusion of Streamlit's user-friendly design and OpenAI's powerful GPT-3.5 model. Here are its highlights:

<ul style='text-align: left;'>
<li><strong>Intuitive Interface</strong>: Built with Streamlit, ChatMate offers a clean, responsive user experience, allowing for natural dialogue with the AI.</li>
<li><strong>Advanced NLP</strong>: Incorporating OpenAI's most advanced GPT model, the app provides nuanced understanding and generation of human-like text, showcasing the model's impressive capabilities.</li>
<li><strong>State Management</strong>: Utilizes <code>ConversationChain</code> and <code>ConversationMemory</code> from <code>langchain</code> to preserve the context and flow, ensuring coherent and engaging interactions.</li>
<li><strong>Python Proficiency</strong>: The app's robust backend, written in Python, reflects the data scientist’s adeptness in programming and system design.</li>
<li><strong>Secure Interaction</strong>: Streamlit's session state management is used for secure API key handling and user input retention across sessions.</li>
</ul>

ChatMate is developed by Lorentz Yeung
""", unsafe_allow_html=True)

#st.session_state['API_Key']= st.sidebar.text_input("Put your OpenAI API Key here please, the system will enter for you automatically.",type="password")

# summarise_button = st.sidebar.button("Summarise the conversation", key="summarise")
#if summarise_button:
#    summarise_placeholder = st.sidebar.write("Nice chatting with you my friend ❤️")



# Function to get response from the model
def getresponse(userInput, api_key):

    if st.session_state['conversation'] is None:

        llm = ChatOpenAI(
            temperature=0,
            openai_api_key=api_key,
            model_name='gpt-3.5-turbo'
        )

        st.session_state['conversation'] = ConversationChain(
            llm=llm,
            verbose=True,
            memory=ConversationSummaryMemory(llm=llm)
        )

    response=st.session_state['conversation'].predict(input=userInput)
    print(st.session_state['conversation'].memory.buffer)
    

    return response



response_container = st.container()
# Here we will have a container for user input text box
container = st.container()

# User input and response display
with container:
    with st.form(key='my_form', clear_on_submit=True):
        user_input = st.text_area("Ask me questions please", key='input', height=100)
        submit_button = st.form_submit_button(label='Send')

        if submit_button:
            st.session_state['messages'].append(user_input)
            model_response=getresponse(user_input,st.session_state['API_Key'])
            st.session_state['messages'].append(model_response)
            

            with response_container:
                for i in range(len(st.session_state['messages'])):
                        if (i % 2) == 0:
                            message(st.session_state['messages'][i], is_user=True, key=str(i) + '_user')
                        else:
                            message(st.session_state['messages'][i], key=str(i) + '_AI')